Reserpine-induced Alterations in the Processing of Proenkephalin in Cultured Chromaffin Cells INCREASED AMIDATION*

We have used antisera directed towards eight differ- ent portions of the proenkephalin molecule to examine the processing rates and patterns of proenkephalin- derived peptides in chromaffin cell cultures in the presence and absence of reserpine. Reserpine treatment produced profound effects on the molecular weight profile of nearly all enkephalin-containing peptides. Increased production of low molecular weight immunoreactive [Met'lenkephalin, [Leu'lenkephalin, [Met'] enkephalin-Ar~-Gly7-Leus, and [Met'lenkephalin-Arg6-Phe7 was observed in reserpine-treated cultures; immunoreactivity corresponding to several intermediate sized enkephalin-containing peptides such as Peptide B and the high molecular weight [Met'len-kephalin-Args-Gly'-Leu8 immunoreactive peptide was decreased. The production of two amidated opioid peptides, amidorphin and metorphamide, was greatly ac- celerated in the presence of reserpine. The increased levels of low molecular weight enkephalins could not be accounted for by assuming decreased basal release. These results indicate that reserpine treatment is able to increase the extent of post-translational processing of proenkephalin within chromaffin cells.

Opioid peptides of the enkephalin family are widely distributed within the central and peripheral nervous system and are thought to participate as neurotransmitters/neuromodulators in many neuronal pathways. The high concentrations of enkephalins found in the adrenal medulla of several species (1) and the presence of circulating enkephalin-immunoreactive peptides (2) are suggestive of a hormonal role for adrenomedullary enkephalins. This role is as yet undefined but may relate to the known cardiovascular effects of synthetic opiates and endogenous opioid peptides (reviewed in Ref.

3).
Wilson et al. (4,5) first demonstrated that reserpine treatment of cultured adrenal chromaffin cells is able to increase levels of radioreceptor-assayable enkephalins, including [Met'lenkephalin and [Leu'lenkephalin. This effect was initially attributed to an increased biosynthesis of proenkephalin (5). However, Eiden et al. (6) and Naranjo et al. (7) have shown that reserpine treatment of chromaffin cell cultures effectively decreases levels of proenkephalin mRNA. Based on their finding of increased amounts of low molecular weight * This study was supported by Biomedical Research Support Grant SO-RR-5376, a starter grant from the Pharmaceutical Manufacturer's Association, and National Institutes of Health Grant AM35199. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. immunoreactive [Met5]enkephalin, Eiden et al. (6) have suggested that reserpine is able to increase the production of [Met'lenkephalin from proenkephalin. Whether the production of other low molecular weight enkephalins is similarly increased by reserpine was not investigated a generalization of this effect to include all of the known processing products of proenkephalin would strongly suggest that reserpine exerts a direct effect on the post-translational processing of proenkephalin. An alternative possibility which must be considered is that reserpine acts to decrease the spontaneous release of low molecular weight opioid peptides from chromaffin cells, thus allowing them to accumulate intracellularly.
In the present study, I have addressed the mechanism by which reserpine can induce changes in the levels of low molecular weight opioid peptides by investigating the posttranslational processing of enkephalin-immunoreactive peptides in chromaffin cell cultures in the presence and absence of reserpine. Eight different antisera, directed toward different portions of the proenkephalin molecule, were used to characterize the molecular weight profile of enkephalins present in the adrenal medulla, in cultured adrenal chromaffin cells, and in cultures treated with reserpine. I have also investigated the influence of reserpine treatment on the release of enkephalin-containing peptides. My results demonstrate that reserpine treatment of chromaffin cell cultures rapidly increases the levels of a variety of smaller proenkephalin-derived peptides, some of which require further posttranslational processing such as amidation.

MATERIALS AND METHODS
Preparation of Chromffin Cell Cultures-Chromaffin cell cultures were prepared from bovine adrenal glands (obtained from a local slaughterhouse) following the method of Wilson et 01. (8) with the following modifications. The initial collagenase digestion of the glands was performed using repeated manual retrograde perfusion with warm 0.5% collagenase (Sigma, Type 1A) in place of mechanical perfusion; fresh collagenase was used each time. DNase (6 mg/ml; Sigma) was included in the collagenase medium and in the Percoll gradient in order to reduce aggregation. The collagenase digestion was terminated by washing the cells three times with Locke's solution containing 2% bovine serum albumin (Fraction V; Miles Laboratories). Following the Percoll gradient step, the dissociated cells were washed three times with Locke's solution and were finally resuspended in a minimal volume of Dulbecco's modified Eagle's medium (Gibco) containing 10% heated fetal calf serum (Gibco), 10 mM HEPES,' 40 mg/liter gentamycin, 10 p~ cytosine arabinoside, 100 units/ml penicillin, and 100 pg/ml streptomycin. An aliquot of the cell suspension was counted using Trypan Blue to determine cell yield and viability. Cells were resuspended in Dulbecco's modified Eagle's medium and plated in collagen-coated multiple 12 or 24 well The abbreviation used is: HEPES, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid. and Processing of Proenkephalin tissue culture dishes (Costar) at a density of 2.5 X lo' cells/cm*. The medium was changed on the third day after plating; reserpine (1 x M, diluted in Dulbecco's modified Eagle's medium) was added at this time. Cultures were homogenized in 0.5 ml of ice-cold extraction buffer (1 N acetic acid containing 20 mM HC1 and 0.1% P-mercaptoethanol) at various time points after the addition of reserpine. In order to compare the profile of immunoreactive enkephalins present in chromaffin cell cultures with the tissue from which they were derived, fresh medullary tissue was dissected free from cortical tissue on ice and homogenized with 10 volumes of ice-cold extraction buffer. Acid extracts from both tissue and cultures were centrifuged at 20,000 X g for 30 min and the supernatant removed and concentrated by lyophilization prior to gel filtration.
The effect of reserpine on the basal release of enkephalins into the medium was examined by removing medium from the cells at varying time points following exposure to reserpine (1 X M) or control medium. Medium samples were then centrifuged and assayed directly for [Met5]-and [Leu']enkephalin; standard curves were also run in medium. The nicotine-stimulated release of enkephalins from chromaffin cell cultures was studied using cultures which had either been treated with reserpine or control medium for 3 days. Replicate wells were exposed to varying concentrations of nicotine in balanced salts solution for a 15-min period at room temperature (9). The balanced salts solution was then removed and stored frozen prior to radioimmunoassay for [Leu'lenkephalin. Radioimmunoassay standards were also run in balanced salts solution.
Gel Filtration-Separation of the various immunoreactive molecular weight forms of proenkephalin-derived peptides was achieved using a 60 X 0.9-cm column of Sephadex G-75, equilibrated, and eluted in 1 N acetic acid containing 0.1 mg/ml crystalline bovine serum albumin (Behring Diagnostics). Columns were run at 4 "C; the flow rate was 2 ml/h and 0.68-ml fractions were collected. The column was standardized using blue dextran, soybean trypsin inhibitor, "'Ilabeled Peptide B, 12'I-[Met']enkephalin-Arg6-Phe7, and cobalt chloride.
Radioimmunoassay and Antisera Specificity-The [Met'lenkephalin radioimmunoassay was carried out using an antiserum generously donated by S. Sabol (National Institutes of Health); the specificity of this antiserum has been described previously (10,11). Briefly, this antiserum is predominantly carboxyl-directed; [Met'lenkephalin-Arg6 shows only a 0.5% cross-reaction, while [Leu'lenkephalin crossreacts by 10%. The details of the [Leu'lenkephalin radioimmunoassay have been previously reported (12). The {Met'lenkephalin-Arg6-Phe7 antiserum was obtained through the courtesy of Drs. J. Schwartz and I. Mocchetti (National Institutes of Mental Health); this antiserum is also primarily carboxyl-directed (13) and exhibits 50% cross-reaction with Peptide B. The [Met'lenkephalin-Arg6-Gly7-Leu' antiserum was raised in this laboratory by immunization of New Zealand White rabbits with [Met'lenkephalin-Arg6-Gly7-LeuR conjugated to succinylated hemocyanin; details of the antiserum production and specificity have been reported (14). Metorphamide antiserum was obtained from E. Weber (Oregon Health Sciences University); the characteristics of this antiserum were previously reported (15). Amidorphin antiserum was provided by B. Seizinger, D. Liebisch, and A. Herz (Max-Planck Institut) and has been described by Seizinger et al. (16), while the characteristics of the synenkephalin antiserum, provided by D. Liston and J. Rossier, were reported by Liston et al. (17). Antiserum to Peptide B was raised in this laboratory in New Zealand White rabbits with synthetic Peptide B coupled to thyroglobulin. The production of this antiserum will be described in a separate paper.' This antiserum shows no detectable cross-reaction with [Met'lenkephalin, [Leu'lenkephalin, or [Met'lenkephalin-Arg6-Gly7-Leu'; [Met'lenkephalin-Arg6-Phe7 cross-reacts by 0.37%.
Radioimmunoassays were carried out in duplicate in a total volume of 0.3 ml radioimmunoassay buffer (0.1 M sodium phosphate buffer, with 50 mM sodium chloride, 0.1% bovine serum albumin, 0.1% pmercaptoethanol, and 0.1% sodium azide, pH 7.4) containing approximately 10,000 cpm of iodinated peptide and antiserum at an appropriate final dilution (ranging from 1:4,000 to 1:50,000). Following overnight incubation at 4 "C, antibody-bound labeled peptide was separated from free labeled peptide using polyethylene glycol with bovine y-globulin as a carrier. More specific details of the radioimmunoassay procedure used have been described recently (18). The interassay coefficients of variation for the various radioimmunoassays ranged between 13 and 20%. Since several of the antisera used cross-' Lindberg, I., and White, L. (1986) Biochem. Biophys. Res. Commun., in press. react with higher molecular weight species, results always refer to peptide immunoreactivity rather than to authentic peptide. Fig. 1 depicts the molecular weight profile of several immunoreactive enkephalins present in extracts prepared from fresh adrenal medullary tissue. Most of the immunoreactive [Met'lenkephalin-Arg6-Phe7 as well as immunoreactive [Met'Jlenkephalin-Arg6-Gly7-Leu' is contained in species with apparent molecular weights between 4 and 8 kDa. In contrast, immunoreactivity corresponding to [Leu'lenkephalin and [Met'lenkephalin was eluted at the position expected for the authentic pentapeptides. This result is most likely due to the fact that these two antisera cross-react poorly with higher molecular weight immunoreactive substances. Fig. 1 also shows the molecular weight profile of amidated opioid peptides derived from proenkephalin. Immunoreactivity corresponding to amidorphin and metorphamide was eluted at positions appropriate to the molecular weight of these peptides. The molecular weight profile of immunoreactive Peptide B, which also elutes at the expected molecular weight position, is also shown in this figure. These results indicate that the major enkephalin-containing peptides in both the adrenal medulla as well as in chromaffin cell cultures are peptides of intermediate molecular weight (Le. larger than the penta-to octapeptides, but more fully processed than proenkephalin).

RESULTS
In Fig. 2 (left panels), the molecular weight profiles of immunoreactive enkephalins present in 3-day-old chromaffin The molecular weight profiles of amidated opioid peptides present in cultured chromaffin cells are shown in the left panels of Fig. 3; both immunoreactive amidorphin as well as metorphamide exhibit elution times characteristic of the authentic peptides. The levels of metorphamide are extremely low in control cultures, but rise dramatically upon treatment of the cultures with reserpine (Fig. 3, right panels). Amidorphin levels also exhibit a pronounced reserpine-induced increase, although the magnitude of this increase is not as great as that of metorphamide.
In Fig. 4, the molecular weight profile of chromaffin cell immunoreactivity corresponding to two intermediate-sized enkephalin-containing peptides, Peptide B and synenkephalin, is shown. Synenkephalin immunoreactivity consists predominantly of two forms, corresponding to apparent masses of approximately 25 and 15 kDa; the position of the lower mass immunoreactive peak corresponds to the elution position of iodinated synenkephalin (not shown). Peptides with Peptide B immunoreactivity elute as a single immunoreactive species in the position of the iodinated Peptide B marker (Fig. 4, left panels). Following treatment of cultures with reserpine, a slight decrease in the amount of the larger synenkephalin-immunoreactive peptide is observed, while no change in the amount of synenkephalin itself is seen. Peptide B-immunoreactive peptides appear to decrease following reserpine treatment (Fig. 4, rightpanels). In general, the profile of the immunoreactive proenkephalin-derived peptides present in chromaffin cell cultures was very similar to that ob- served in the parent tissue, implying that chromaffin cells represent an appropriate model in which to study the biosynthesis of enkephalins.
The time course of reserpine-induced changes in low molecular weight peptide concentration was assessed by measuring levels of three opioid peptides at varying times following the administration of reserpine t o the cultures (Fig. 5). These peptides were chosen for assay because the previous experiments showed that they consisted predominantly if not exclusively of one molecular weight form. As shown in the top panel of Fig. 5, the levels of immunoreactive metorphamide increase steadily after exposure of the cells to metorphamide (dashed line); control cells maintain constant levels of this peptide (solid line). In contrast, the time course of activation of amidorphin production (middle panel) indicates that this peptide does not appear to increase as much as metorphamide. Unlike metorphamide, increased amidorphin production exhibits a plateau at about 12-24 h. Similarly, [Leu'lenkephalin production, shown in the bottom panel, also does not respond to reserpine as extensively as does metorphamide production. As observed in the case of amidorphin, reserpine-stimulated [Leu'lenkephalin production also exhibits a plateau at about 12-24 h. Some experiment to experiment variability in the degree of stimulation by reserpine of low molecular weight enkephalins was observed. The reasons underlying this variability are not known but may have to do with the fact that reserpine effects appear to be highly dependent on drug/tissue ratios (19). It is thus possible that variations in cell plating and/or fibroblast content may have contributed to a variable effectiveness of the drug. However, it is of interest to note that metorphamide production always showed a far greater increase in response t o reserpine than amidorphin or [Leu'] enkephalin (e.g. in five separate experiments, the increase in metorphamide levels ranged from 800 to 1300%, while the increase in [Leu'lenkephalin and amidorphin production ranged from 50 to 400%).
The possibility that the reserpine-induced increase in enkephalin production observed after reserpine treatment was due directly to decreased release of Iow molecular weight peptides was tested by examining the levels of [Met'lenkeph-  alin in media from cultures which had been exposed to reserpine or control media for varying amounts of time. As may be seen in Table I, exposure of cells to reserpine significantly reduced the basal release of [Met'lenkephalin into the medium. However, at 3 days, these media concentrations represent only approximately 6% and 4% of the cellular levels of [Met'lenkephalin in the control and reserpine-treated samples, respectively. Since cellular low molecular weight enkephalin levels were stimulated between 2-to %fold by reserpine treatment it appears unlikely that inhibition of basal release can be solely responsible for the increased levels of enkephalins observed during reserpine treatment. The effect of acute or chronic treatment with reserpine on the nicotine-stimulated release of enkephalins was investigated in a separate experiment (shown in Fig. 6). Chromaffin cells cultured in the presence of reserpine released greater quantities of immunoreactive [Leu'lenkephalin in response to nicotine than did cultures never exposed to reserpine. However, the presence of reserpine during the release experiment did not affect the nicotine-stimulated release of [Leu'lenkephalin (Fig. 6). Taken together with the findings presented in Table I, these results argue against the notion that the primary action of reserpine in increasing cellular levels of low molecular weight enkephalins is to inhibit t,he release of these peptides.

TABLE I Effect of reserpine on basal release of [MePIenkephalin into the medium
Ascorbate has been reported to be an essential cofactor in the amidation of a-melanotropin (20, 21). It was therefore of interest to examine whether the reserpine-induced stimulation of metorphamide and amidorphin, both amidated peptides, also exhibits a requirement for added ascorbate. Treatment of cultures was carried out in the presence and absence of ascorbate (250 p~) .
In agreement with the results of Wilson The open circles represent data obtained from cells acutely exposed to reserpine (concurrently with nicotine), while the closed circles represent cells exposed only to nicotine. Leu-ENK, [Le~~lenkephalin.  and Kirshner (9), ascorbate supplementation was found to have no effect on the basal levels of opioid peptides; ascorbate also had no effect on the reserpine-induced increase in metorphamide or amidorphin (Table 11).

DISCUSSION
The above data indicate that reserpine is able to increase the production of not only [Met'lenkephalin, but also [Leu6] enkephalin, [MePIenkephalin-Arg6-Phe7, and [Met'lenkephalin-ArgG-Gly7-Leus. Reserpine thus appears to accelerate the generation of all low molecular weight enkephalins; however, the levels of intermediate-sized enkephalin-containing peptides, such as Peptide B and the 5.3-kDa fragment of proenkephalin, are decreased in response to reserpine treatment. These results suggest that reserpine is able to increase the general activity of proteolytic processing enzymes, perhaps within the chromaffin granule. In order to ascertain whether activation of amidation, another important post-translational processing event, can also occur in the presence of reserpine, we examined the effects of reserpine treatment on the production of two amidated opioid peptides derived from proenkephalin. The levels of both amidated enkephalins, but in particular metorphamide, were observed to increase rapidly in response to reserpine treatment, suggesting that amidation is indeed activated in the presence of reserpine. This effect may be due to stimulation of amidating enzymes by reserpine; Hook et al. (22) have reported that certain kinetic parameters of the carboxypepotidase B-like processing enzyme are altered in response to reserpine treatment. Alternatively, it may be speculated that amidation is not normally a rate-limiting processing step; when reserpine acts to increase the amounts of glycine-extended precursors to amidated peptides, these precursors then become rapidly amidated. Unlike the production of a-melanotropin in intermediate pituitary cell cultures (20,21), the production of amidated proenkephalin-derived peptides in chromaffin cell cultures was not affected by the inclusion of ascorbate in the medium. These results are surprising in view of the rapid loss of endogenous ascorbate from chromaffin cells placed in culture (23) and imply that either the existing ascorbate concentration is sufficient to maintain amidation, or that other reducing equivalents are utilized for this reaction (24).
Interestingly, the time course of reserpine-induced effects on the production of amidated peptides suggests that metorphamide production continues to increase long after the production of [Leu'lenkephalin has reached a plateau. These results might reflect the sequential nature of the processing steps required to generate these peptides from their presumed common precursor, Peptide E (which contains the sequence of metorphamide at its amino end and which terminates in [Leu'lenkephalin (25)). Pulse-chase studies of enkephalin production will be required to demonstrate the precise effects of reserpine on this step of the enkephalin biosynthetic pathway; such experiments are now in progress.
The inhibition of basal release observed in the presence of reserpine, while not in itself sufficient to explain the large increases in intracellular low molecular weight enkephalins, represents an intriguing phenomenon. The data presented above as to the effects of nicotine on the release of enkephalins from reserpine-treated cells indicate that reserpine does increase enkephalins in a pool which is stimulus secretion coupled. At the same time, reserpine apparently paradoxically lowers the basal release of enkephalins into the medium. It is interesting to note that forskolin and cyclic AMP, agents which also increase intracellular enkephalin levels (but act by increasing the transcription of proenkephalin mRNA) both enhance rather than decrease basal release of enkephalins into the medium (26, 27).
The mechanism of action of reserpine in increasing the post-translational processing of intermediate sized enkephalin-containing peptides remains elusive. The subcellular loci of the post-translational processing events involved in the generation of low molecular weight enkephalins have not yet been determined. However, in other prohormone systems, late proteolytic processing steps as well as other processing events (such as removal of carboxyl-terminal basic residues, acetylation, and amidation) are thought to take place within the secretory granule (reviewed in Refs. 28 and 29). In addition, reserpine is known to bind to the chromaffin granule membrane, where it acts to inhibit catecholamine transport into the granule by blocking the amine translocator (reviewed by Stitzel (30); see also Ref. 19). Assuming that the locus of action of the drug is intragranular, it may be speculated that decreased intragranular concentrations of catecholamines are involved in the reserpine-induced activation of post-translational processing. This effect may be a purely physical phenomenon, such as an improved intragranular milieu for proteolytic processing enzymes resulting from decreased catecholamine content. Alternatively, it is possible that binding of reserpine to chromaffin granules induces conformational alterations in granule membrane structure which then result in increased activity of intragranular processing enzymes. Further research will be necessary to distinguish between these possibilities. Note Added in Proof-Similar results have recently been reported by Eiden, L. E., and Zamir, N. (1986) (J. Neurochem. 5,1651-1654).